All posts by John.Dawes@marketingscience.info

A Branding lesson – Wedgwood, 1770’s.

Josiah Wedgwood was an entrepreneurial potter and business owner, the founder of the famous brand of porcelain.  This is a slice of his brand-building story:

Brand Marketing was virtually unheard of in the mid-eighteenth century.  Only a handful of luxury goods, such as Chippendale furniture or Meissen porcelain were known by their manufacturer’s names.  Until about 1770, most potters did not mark their products.  The few earthenware and porcelain manufacturers that did, such as the Chelsea porcelain factory, generally used signs, symbols or the location of the factory as identifying marks.  Wedgwood changed this practice in the late 1760’s by impressing his own name in the unfired clay.  His works were thus much less vulnerable to forgery than other makers, and, as Josiah understood, every piece advertised the Wedgwood name.  By 1772, everything made at Wedgwood’s pottery, useful or ornamental, carried his name.

And here is a snippet of his thinking on buyer behaviour:

Josiah realised that many Britons now had more money to spend on nonessential or luxury goods than had their counterparts in previous generations. He also understood that much of this spending was directed toward social emulation.  Eighteenth-century Britons, like modern consumers all over the world, tended to put their money where their aspirations were.  They spent as the rich did, or at least as the income class directly above them.  Wedgwood knew that the middling ranks wanted to ape their social betters, and he planned his sales strategy accordingly.  Josiah actively sought aristocratic and noble commissions and the explicit and implicit endorsements that accompanied their sales.

Once he had completed an aristocratic sale, Josiah lost no time advertising it to a much larger, more profitable market.   Wedgwood took out ads in London newspapers to celebrate his royal patronage.   Competitors … followed Wedgwood’s marketing lead.  These manufacturers used newspaper advertising and urban showrooms to sell their wares.  Some sought aristocratic endorsements.  But none deployed a marketing strategy that rivalled the scope, effectiveness and sustainability of Wedgwood’s.

from  “Josiah Wedgwood, 1730-1795” – a chapter in “Brand New,” by Nancy F Koehn, Harvard Business School Press. 

Gift Cards / Promotion Credit apparently heighten perceived discount

Price promotions occur all the time and often simply involve a straight price cut.  But sometimes the price reduction is in the form of a discount on a later purchase.  For example a promotion for a $1,000 laptop might be a $100 gift card or voucher to use on another purchase later.  Amazon offers a lot of deals in this way, offering “promotional credit”.

These sorts of price promotions arguably might have dual effects.  The customer feels they got a discount on the original purchase.  But then they use the gift card, voucher or promotional credit later – perhaps they feel they’ve received discount again!

Researchers Cheng and Cruyder (JMR 2018) researched this issue in a series of well-conducted experiments.  First they compared total purchasing across two occasions among (a) people who only got an immediate discount (b) people who got promotional credit of an equivalent amount, to use on a later purchase.  Total purchasing was higher in the promotion credit group.  Next they compared (in-survey, with no actual money spent) purchasing among people who got no discount, discount, or promotional credit.  Purchasing was about 20% higher in the promotional credit group.  Further testing suggested the explanation was that consumers given promotional credit feel as if they spend less in total across two purchases than those only given an (equivalent) discount.   Additional experiments suggested promotional-credit type deals work better than straight discounts, mail-in rebates or cash-backs, because the promotional credit is more easily linked to a subsequent purchase and there is a heightened tendency to feel that two discounts have been received. In total the study ran six experiments with convergent results.

The main “take-out” is that when consumers receive a Gift Card or promotional credit to use on a subsequent purchase, they feel like they are spending less over the two purchases than they would if they simply received a price discount on the first purchase.  The end result is that promotional credits tend to result in higher total spending (over two purchases) than straight discounts.

 

Reference:

ANDONG CHENG and CYNTHIA CRYDER (2018) Double Mental Discounting: When a Single Price Promotion Feels Twice as Nice Journal of Marketing Research, April p. 226-238.

The ‘near impossibility’ of measuring Advertising ROI

Summary: 25 large-scale experiments with over 2 million households in co-operation with a large retailer, a stockbroker, and Yahoo show that it is extremely difficult to identify advertising ROI.  Why? Because there is such massive cross-sectional variation across households in spending; plus huge variation in spending by the same household across time; and huge variation in purchase timing (sometimes a household buys twice in two weeks, sometimes only once in a year or not at all) – this massive variation makes it almost impossible to distinguish advertising effects from random noise.  This is the case even when one has individual household data that matches both ad exposure and purchasing.

The market context

Here is some of the detail of the market and advertising context.  I have simplified & explained the detail of the spending that is in the original study, to make it clearer to readers of this post.  The context is a retailer, that can target ads to specific households. Furthermore it knows the exact amount of money those hosueholds spend with it.  The study also uses a stockbroking firm that advertises to consumers with more or less the same sort of figures as below.

Typical scenario: An advertiser is going to hit households with approximately 35 display ads in a 2-week campaign period.  The advertised product has a gross margin of 50%.  The cost of the advertising per household is 14 cents (based on a price of $4 per thousand people, per ad = .4 cents per ad x 35 = 14 cents) for the campaign.

The average sales per consumer is $7 in the time period, but the standard deviation (i.e. the variation in sales across people) is $75.   This means households vary from $0 to hundreds of dollars in purchases over the expected duration of the campaign.

Next, the advertising ROI goal is a 25% ROI.  Spending 14 cents per person and getting a 25% ROI means the goal is to generate 14 x 1.25 = 17.5 cents profit per exposed household.  In turn, this means we need to generate 35 cents extra sales revenue per household, on average.  This figure comes from the assumption of a 50% gross margin, since 50% margin on 35 cents of sales is 17.5 cents, which is in turn 25% larger than the 14 cents we spend hitting each household with ads.

The advertiser then selects a control group which will not see the advertising, and a treatment group that will.  And remember, it’s not as if the control group is quiet / stable in the campaign – an awful lot of unexposed households will buy the product. 

So this selection of control and treatment groups, and hitting the treatment households with ads is what the researchers did, in 25 different experiments, with variations on these basic figures.  To verify if a campaign reached its 25% ROI, though, they had to detect an average difference of 35 cents or more per household in sales, between a treatment and control group, when average sales per household are $7 and the standard deviation in sales across households is $75.  This is just too small a difference to detect, given the massive variation in household spending.

Conclusion

Rao and Lewis concluded even with treatment groups of 200,000 consumers, these real-world experiments were quite underpowered to be able to reasonably verify if the campaigns reached the ROI target, or even if they had any effect at all.  You might think, maybe it was because the target ROI was quite small – what if it was 50%, not 25%?   The answer is it would not make much difference.  It would mean finding a sales difference of 40 cents per household rather than 35, again with an average sales level of $7 and a standard deviation of $75.  Again, the advertising effect on sales would be trivially small compared to the natural variation that occurs.

Management take-out

So the take-out is, it’s actually very difficult to measure advertising ROI – even with carefully controlled experiments – due to massive variation in baseline spending levels across households.  Or as the researchers concluded themselves, “We find that even when ad delivery and consumer purchases can be measured at the individual level, linked across purchasing domains, and randomized to ensure exogenous exposure, forming reliable estimates on the returns to advertising is exceedingly difficult, even with millions of observations”.

Based on this very large study, managers should be very cautious about promises from vendors to calculate their ROI, or that guarantee a certain ROI on advertising.

The full study (lots of maths if you like that sort of thing) is at http://justinmrao.com/lewis_rao_nearimpossibility.pdf

 

 

Is Price really the most powerful profit lever?

Answer: Only if you assume you can put price up and nothing happens to unit sales, which is unlikely.

Pricing
Powerhouse consultants McKinsey state that “Pricing right is the fastest and most effective way for managers to increase profits. Consider the average income statement of an S&P 500 company: a price rise of 1 percent, if volumes remained stable, would generate an 8 percent increase in operating profits (Exhibit 1)—an impact nearly 50 percent greater than that of a 1 percent fall in variable costs such as materials and direct labor and more than three times greater than the impact of a 1 percent increase in volume” – source, http://www.mckinsey.com/insights/marketing_sales/the_power_of_pricing.
This storyline has been around for a long time. Bain said the same sort of thing years ago. More recently the columnist Mark Ritson publicised the idea, and there are dozens of consultant web sites echoing this line.

So, getting price up apparently works better on bottom-line profits than either improving sales volume, or reducing costs – percent for percent. Is this true, and if so, how is it possible ?

The answer is yes it is true only IF you can increase price without sales dropping off. That’s a very big IF. We’ll take a look later in this piece at the evidence about what actually happens to brands when they increase price. Second, this statement assumes the effort involved to get price up is cost neutral. That’s another pretty big caveat, in fact it’s pretty unrealistic. Surely you would not rush in and just blindly put prices up on everything, so there would be time, therefore expenses involved in such a process.

The other important point is that the statement ‘price has more effect on bottom line profit than volume or costs’ is that it’s in theory true, but true by definition. It’s simply a function of margin arithmetic. I’m indebted to the late John Scriven, formerly of London SouthBank for explaining this to me years ago, and it took a couple of years for the explanation to stick.

Explanation
In basic terms, the product you sell always has some variable and fixed costs. And those costs – per item – are only a fraction of your selling price. So if you reduce costs it only applies to a fraction of your selling price. And if you increase unit sales, you only retain a proportion of the additional margin, because you’ve got some cost component on every item you sell. But if you increase price, the increase applies to ‘all of your price’.  If that’s clear, stop reading now, otherwise here is an example to illustrate.

Example
You make ice creams for 50c and sell them for $1. Contribution, or margin on each one is 50c.  You sell 1,000 ice creams.
If you reduce your costs by 1% you save half of one cent per ice cream, which adds 1000 x half a cent = $5 extra profit.
If you sell 1% more ice creams you sell 1000 x 1% = 10 more ice creams and keep 50c margin on each, = $5 extra profit.
But if you increase your price by 1% it’s 1% on a dollar, not 1% of 50 cents. So the extra on the bottom line is 1000 x 1 cent = $10.

But, be sure to note that the assumption, as before, is that your prices have increased by 1% AND your sales have remained exactly as before.

So McKinsey (& Bain & Ritson) are in theory correct, if you accept you can raise prices with no loss of sales. It’s theoretically correct, because of the simple arithmetic linking cost, prices, volumes and profits. Whether you can get prices up with NO loss of sales, and if it’s easier or harder than getting costs down or volume up, is another matter. The McKinseys and Bains will say, you might be giving away unnecessary discounts and they can help you to remedy that. But doing so will probably involve a platoon of consultants and cost a million, so the idea you can get prices up easily or at no cost quickly disappears. And if it were possible to get prices up one percent with no loss, why not two percent or ten percent? Where does it stop?
Let’s now look at real-world price changes.

The Real World – it gets more complicated
In the real world if you increase price your unit sales go down (just as is equally well known that price drops increase sales volume). Another way of saying this that brands are price-elastic. Price elasticity is the change in unit sales for every 1% price change. On average brand price elasticity is -2.5, that is, if price goes up 1% then sales go down 2.5%. This is no mere theoretical concept. It is the result of dozens of studies including several from our own Ehrenberg-Bass researchers. And it varies for big and small brands and according to other factors like competitor prices, signalling and price passing. However, there is no evidence that price elasticity is related to amorphous concepts like brand equity or brand strength (over and above the effect of the brand’s size in market share).

So whether you make more money from putting prices up depends on how price-elastic your brand is, and what its margin is before the price increase. If you have a low margin brand you can gain a lot from even a small price increase, because the proportional gain in margin is huge (e.g. at an extreme, perhaps the price increase could double your margin from previously). But if you have a brand with a high margin already, a price increase that knocks sales down even by a small amount will drop your total margin / contribution. Why? Because you already make so much money from every item sold, that if you sell a few units less you are losing a lot of margin in total.

In summary:

Statements that say price is the most powerful profit lever assume you can raise prices with no impact on unit sales. But ample evidence shows unit sales fall when you raise prices.

On average, brand price elasticity is around -2.5, so every 1% price increase relative to competitors drops unit sales by 2.5%

You have potentially more to gain from a price increase on a brand if its margins are currently low. But you can potentially lose a lot of profit if you raise prices on a brand that already has high margins.

 

 

Millennials’ favourite brands are the same as everyone else’s

Quick summary
Millennials’ favourite brands when surveyed are:  #1 Apple, #2 Nike, #3 Samsung, #4 Target, #5 Amazon, # 6 Sony, #7 Wal-Mart, #8 Microsoft … (etc).  In other words, big/popular brands.  Like the ones we all buy.

Introduction
We all keep hearing about how different Millennials are to other generations.  And that they’re a super important segment, because of their large spending power.  Many firms are apparently creating new brands to accommodate them – for example, Air France created Joon.  Apparently, Millennials want ‘more meaning’ from brands, and this is the way to do it.  But if Millennials really are a segment, then they really should have different brand preferences than other people.   What’s the evidence?

Survey
A large survey of US Millennials about their favourite brands was conducted by ad agency Moosylvania, and the results were published recently in Business Insider.   Right at the start of the report, you’re struck by the wide age range of this supposed group:- 17 to 37 years old.  Who in their right mind would stereotype such a diverse group as a market segment?  And that’s just diversity in age.

Results
Anyway, the favourite brands of Millennials were ranked in this report from 100 to 1, with 1 being the most favoured.  Now a common story about Millennials is that they resist or reject older, established brands.  If that were true you would think we would see unfamiliar brands popping up among their favourites.  But we don’t.  These are the top 10 favourite brands among Millennials.

1 Apple
2 Nike
3 Samsung
4 Target
5 Amazon
6 Sony
7 Wal-Mart
8 Microsoft
9 Coca-Cola
10 Google

Leaving aside the fact many of these are technology brands (which always tend to come up high in brand popularity surveys among any population), these are all (big) well-established brands!  There are no brand-new upstarts.

A common theme in the report when trying to explain the brands’ popularity with Millennials is, ‘lots of Millennials buy this brand’. Which is a circular argument, and furthermore misses the point that lots of EVERY demographic buy these brands – that’s why they’re big.

Some other rationalizations are quite amusing in the report.  For example, Van’s is popular in this group “because it [Van’s] went mainstream” and Kraft is hot with Millennials because – wait for it – it got rid of artificial colourings in its macaroni and cheese.

Moral of the story: Millennials most favourite brands are big, established brands. This gives no support to the idea that corporations must create new brands to accommodate Millennials.

US TV – recent stats show high primetime viewing

From a recently published US study using set-top boxes:

“85% of households watch prime-time TV every night, on average watching 88% of the prime-time hours. Thus, any gains from targeting largely arise from what a viewer watches rather than whether they watch. Once a viewing session commences, 30% of households sample multiple shows before selecting one to view, suggesting that show sampling is informative about viewing preferences. Once a show is selected, 60% of viewers watch a show to its conclusion, indicating ad exposure is common once a show is selected. Within a show, we find that viewers’ advertising avoidance is more common when the show is recorded (79% [when recorded] vs 15% [when not recorded]).

from Deng, Y.  & Mela, C. “TV Viewing and Advertising Targeting” Journal of Marketing Research 2018.

  • the text in [ ] brackets I added for clarity.

Beware of using ‘Personas’

Many marketing people are enthusiastic about Personas.  The basic problem of personas is that almost all sales (your sales, the sales of the product class) come from buyers who are not really anything like the Personas.
Take this example from something I read yesterday.

” ….. a health foods company targets moms with college degrees who are health conscious, but they create a strategy for Molly, who is 36, married with two children, has a Master’s Degree, earns six figures, loves to run and only buys organic food for her family.

As the CMO, I would be much more confident targeting the Mollys of the world with a message that our foods are healthy for both her and her family. And although they’re organic and cost more than non-organic foods, she can rest easy knowing that what she (and her family) put in their bodies is good for them. And you can’t put a price on that. “

OK, you’re a health foods company.  Your first mistake would be to think buyers have to be into healthy lifestyles to buy health foods.  That’d be at least worth checking – I’d put money on the fact lots of not-particularly health-conscious people buy healthy or organic cereal etc. at least sometimes.  Your second mistake would be to think the buyers of your product have really sharply defined things in common – moms with college degrees?  Really – so single men, moms without college, women without children or with adult children don’t buy health foods?

The third mistake is to double down on your initial errors and think your strategy should be to target a ridiculously tight persona – 36 years old (not 30, not 35 even – and not even 30 to 40 which would still be unnecessarily restrictive); married with two children, not one or three or none; Master’s degree – why is this needed?  Six figure income – why?  Loves to run – why?  Going to the gym, or playing sport isn’t good enough, Molly has to love to run and ONLY buy organic food.  This description would match perhaps one in fifty thousand households if you were lucky.  So following this strategy would mean you ignore 99.9+ percent of people.  The final flaw is thinking you could actually find and selectively target & reach such people anyway.

Marketers like the idea of personas because they seemingly make the job of dreaming up content easier – “what would Molly like?”.  But you run the real danger of creating irrelevant content / ad creative and trying to only talk to a minority – because Molly only exists in your imagination, and 99%+ of people buying the category you compete in are not like Molly.

Advertising wisdom from 1917

This statement was written over 100 years ago, but still applies today.  Brand advertising, that is consistent over time, and prominently features the brand’s distinctive elements, builds buyer recognition and acceptance – which is financially valuable to the brand owner.  Not only valuable … long-term valuable.  But the work has to continue – you have to keep doing it.